Automatic target recognition (ATR) in artificial systems can involve complex computational models and data processing. To perform ATR, artificial systems have been modeled after the human brain. However, computational models based on neurobiology and physiology of human brain may not be suitable for performing certain ill-defined tasks such as pattern recognition, sensor fusion, fault-tolerance control, adaptation to environment, etc.